ECOLOGICAL SOUNDING © 2003 Blackwell Publishing Ltd. http://www.blackwellpublishing.com/journals/geb Global Ecology & Biogeography (2003) 12, 1–3 Blackwell Publishing, Ltd Island ecology and contingent theory: the role of spatial scale and taxonomic bias ANDRÁS BÁLDI*‡ and DUNCAN MCCOLLIN† *Animal Ecology Research Group of the Hungarian Academy of Sciences, Hungarian Natural History Museum, Ludovika tér 2, Budapest, Hun- gary, H-1083, E-mail: baldi@ludovika.nhmus.hu, †Landscape and Biodiversity Research Group, School of Environmental Science, University College Northampton, Park Campus, Northampton NN2 7AL, U.K., E-mail: duncan.mccollin@northampton.ac.uk ABSTRACT Scale, the scale dependency of patterns and processes, and the ways that organisms scale their responses to these patterns and processes are central to island and landscape ecology. Here, we take a database of studies in island ecology and investigate how studies have changed over a 40-year period with respect to spatial scale and organisms studied. We demonstrate that there have been changes in the spatial scale of islands studied and that there is taxonomic bias in favour of vertebrates in island ecological studies when compared to scientific publications as a whole. We discuss how such taxonomic bias may have arisen and discuss the implications for ecology and biogeography. Key words invertebrates, island area, island biogeography, island ecology, plants, spatial scale, vertebrates. Ecological patterns and the laws, rules and mechanisms that underpin them are contingent on the organisms involved and their environment (Lawton, 1999, p. 177). There is an increasing recognition of the importance of scale and the scale dependency of patterns and processes in ecology, and the ways that organisms scale their responses to these patterns and processes (Levin, 1992; Wiens et al ., 1993; Hubbell, 2001). Here, we investigate how studies in island ecology have changed over a 40-year period with respect to spatial scale and the organisms studied. Island ecology has a long pedigree in ecology and biogeography, stretching at least as far back as the work of Alfred Russel Wallace (1880), although in recent years island ecology has been dominated by island biogeography sensu MacArthur & Wilson (1967) (Whittaker, 1998). The MacArthur–Wilson model is based on the assumption that the numbers of species found on islands is a balance between the opposing rates of immigration and extinction, the latter influenced by island area acting on population size and the former influenced by distance from the mainland. As such, a common methodological approach has been to count numbers of species on a sample of islands (oceanic or habitat) of differing sizes and relate patterns of species richness to geographical (e.g. area, isolation) or physical (e.g. habitat diversity, altitude) attributes of islands (e.g. McCollin, 1993; Báldi & Kisbenedek, 2000). Alternative approaches, using more intensive studies on a smaller range of islands (e.g. Blondel et al ., 1999), can also be very inform- ative but tend to be much less common. We ask whether there are trends in the spatial scales chosen by researchers working in island ecology over time and whether there are trends in taxa studied. If any such trends exist we ask what are their causes and query whether they are exclusive to island ecology or whether they reflect a more fundamental bias or lack of rigour in ecological method. These are important questions not just for island ecology, but potentially have wider implications for landscape ecology and biogeography. We use a pre-existing dataset assembled by Wright et al. (1998) for their analysis of nestedness in ecological commun- ities. The spread of papers in this dataset is not comprehen- sive, but it does represent a large sample of studies published between 1933 and 1993. The 163 independent data matrices used by Wright et al. (1998) include useful information for such an analysis including taxon, location and island size, among others (see www.aics-research.com/nested). In order to test this dataset for trends with respect to spatial scale, data for numbers of species, numbers of islands, minimum and maximum area were entered into correlation ‡ Corresponding author.